On the basis of data in another context, not controlled for age, sex or race, cancer patients with well-established disease showed very high error scores on the Bender-Gestalt perceptual-motor test. In a forced blind setting the test will be given to pre-biopsy patients. After matching for sex and race, the scores will be analyzed, correcting for age by covariance methods, for ability to distinguish cancer and non-cancer patients.